°Spond, as 
in the Y7 
1e also in 
e between 
lon on the 
in the two 
jections of 
secondary 
S, the pro- 
G Sec g, o 
ect to the 
) B on the 
vill change 
ccentricity 
T and x" 
eckings, is 
vere found 
ights, and 
1e rod and 
1 that em- 
' are made 
b and AB 
/ (Fig. 20). 
‚tion plane 
lination of 
pond iden- 
| in the XZ 
  
Identically, the intersection points b' and a' of the short rod with the plate 
plane must lie on a straight line also passing by M. These four points do not 
longer mutually correspond, i. e. in passing from a' to b’ the length of the short 
rod changes by dS. This change results from the following relation: 
7 dS" = 2 .y.S'.tan © (sec — 1) 
where S' is the length of the short rod in its inclined position. 
As in the previous instances, the measurement of 5 and Y was taken in 
such a way as to eliminate possible height variations of the reference plane. 
The obtained experimental values of the angular eccentricities are, as 
far as the left camera 1s concerned, as follows: 
c — angle between the rod and short rod in the plane YZ — 31" 
y = angle between the rod and short rod in the plane XZ — 14" 
i 
The approximation of these values is of about 4 to 5". 
In closing this analysis of eccentricities it is advisable to draw attention 
again on the remark stated above; viz. that, even though the theoretical 
influence of these eccentricities may seem relevant one should however bear 
in mind what has been said on page 21 and that the remaining small errors 
are practicallv offset by slight variations of the angular orientation elements 
of the cameras (*). 
In particular, it is easy to realize that the high eccentricity b of the short 
rod causes a deformation of the grid projections very similar with that due to 
variation of cross inclination c, and that during the search for the instrument 
zeros, a large part of the deformations is offset and averaged by means of 
this angular element so as to reduce them within satisfactory limits. 
5) Bendings between rod and short rod. — The last condition for the cor- 
rect operation of the projection organ which remains to be considered is the 
rigidity in the connection between the rod and the short rod. 
This element is very important in the Stereocartograph because the or- 
gan is split into two parts. The lack of rigidity causes a variation of the an- 
gle between rod and short rod, which depends upon the inclination of the rod 
as well as upon the direction of this inclination. From this, dangerous discre- 
pancies arise between the pencil described by the short rod and the rod. 
Appropriately, therefore, Santoni provided the projection mechanism 
with a patented « Antiflex » system by means of which the mechanism instead 
of being merely fixed in the centre becomes fixed in the centre and supported 
in two peripheral points. 
The results of the last mentioned checking as well as other details of this 
analysis and a synthetic study of projection errors appear in the final paper, 
which will be published on an Italian magazine. 
We think it advisable, however, to give here provisionally the values of 
the mean projection errors obtained from tests conducted during plotting. 
We, = Lt Gp; wh, = dy; My = 0,08% - 
V 
Ea ; ; 
Planimetric errors are transferred on the plate plane, altimetric errors 
4e equal both with inward and outward base. 
i () Dr. Santoni wishes to point out that really some eccentricities are introduced as instrumental 
adjusteme 
nts with the aim to get lower errors of projection. 
  
  
  
  
  
  
  
  
  
  
  
  
  
Weir ere rte 
  
A 
«